The last Study IV in this thesis work aimed to examine the role of macrophage polarization in TB infection. Since Study I and III were designed to explore the antimicrobial effects of PBA+VitD3, we decided to continue to study the effects of particularly VitD3, on macrophage polarization. In vivo, macrophages can adopt a variety of functional phenotypes depending on subtle and continuous changes in the tissue microenvironment (Mantovani, Sica et al. 2004, Mantovani, Biswas et al. 2013, Italiani and Boraschi 2014).
To define the complexity and plasticity of macrophages, the concept of M1/M2 polarization of macrophage functions has been founded as a simplified conceptual framework describing a non-linear continuum of diverse functional states (Mosser and Edwards 2008).
Similar to Th1 and Th2 cells, classically activated M1 cells are inflammatory while alternatively activated M2 cells are considered wound healing or tissue remodeling macrophages (Mantovani, Sica et al. 2004, Mantovani, Biswas et al. 2013, Italiani and Boraschi 2014). Classically activated M1 macrophages are regarded as key effector cells in the elimination of microbes and cancer cells, while M2 macrophages contribute to the production of extracellular matrix components such as MMPs involved in tissue remodeling (Mantovani, Biswas et al. 2013).
In Study IV, we wished to investigate the ability of VitD3-polarized macrophages to control intracellular Mtb infection compared with polarization of conventional M1 and M2 macrophage subsets, and the phenotypic alterations associated with immune polarization in vitro. Also in this study, we took advantage of our in vitro macrophage infection model, and here we applied a protocol for macrophage polarization with different stimuli as described in in figure 13.
Figure 13. Cytokine and chemokine expression of polarized macrophages upon H37Rv infection.
3.4.2 In vitro polarized macrophages express typical M1 and M2 phenotypic markers but adopt a mixed M1/M2 phenotype post-Mtb infection
Initially, we wanted to confirm that uninfected MDMs activated with classical M1 or M2 stimuli could express phenotype markers typically associated with M1 or M2 polarization and how these phenotypes differed to VitD3-polarized macrophages. An array of different markers used to identify and characterize M1 or M2 macrophages have been described (Macrophage Polarization Mini-review, Bio-rad). We tested different markers using multicolor flow cytometry, and ultimately decided on one M1 and one M2 panel that showed reproducible and reliable results. While VitD3-polarized cells maintained a phenotype similar to unstimulated MDMs, we could determine that M1 markers such as CCR7, CD64, CD86 and TLR2 or M2 markers such as CD163, CD200R and CD206, were proportionately higher on M1- or M2-polarized cell subsets, respectively. M1/M2 polarization of uninfected MDMs
VitD GMCSF GMCSF+LPS+IFNγ MCSF MCSF+IL4
VitD3!polarized M1! like M1!polarized M2#!like M2!polarized Mtb$infection
M1 cells and more rounded appearance of M2 cells. Interestingly, Mtb infection resulted in an up-regulation of most M1 and M2 markers on the different macrophage subsets, generating a mixed M1/M2 profile. Mtb-infected VitD3-polarized macrophages had a significantly up-regulated expression of HLADR, CD86, CD80 and CD206. Both M1 and M2 subsets exhibited an up-regulated CD86 expression, while the M1 subsets also had a significantly up-regulated expression of the M2 marker CD163. A mixed M1/M2 activation profile has previously been shown in several chronic conditions such as melanoma (Bardi, Smith et al. 2018) and multiple sclerosis (Vogel, Vereyken et al. 2013), providing examples where macrophages adopt a mixed phenotype in vivo. As an alternative to mixed M1/M2 responses, there is evidence from in vitro studies that macrophages also can switch their phenotype from M1 to M2 or vice versa depending on sequential micro-environmental challenges in the tissue (Italiani and Boraschi 2014).
3.4.3 Enhanced Mtb growth in both M1 and M2 subsets at 72 h post-Mtb infection compared with VitD3-polarized macrophages despite lower bacterial uptake in M1 cells at 4 h
To study Mtb uptake as well as productive infection and Mtb growth, we infected the different macrophage subsets at 4 h, 24 h and 72 h. Mtb uptake (4 h) and productive infection (24 and 72 h) was monitored using flow cytometric analyses of Mtb-GFP expression in the macrophages, whereas intracellular Mtb growth was assessed using CFU counts after 4 and 72 h post-Mtb infection. Mtb uptake was remarkably lower in M1-like and M1-polarized subsets compared to all other subsets, but especially compared with the M2 subsets (18%
versus 82%). Also intracellular Mtb growth was considerably lower in the M1 subsets compared to the M2 subsets at 4 h post-Mtb infection. But productive infection and Mtb growth inside the M1-like and M1-polarized macrophages progressed rapidly overtime, and at 72 h post-infection, Mtb growth was similar in the M1 and M2 subsets. Notably, while M1 cells lost their capacity to control Mtb growth, M2 cells stabilized and reduced Mtb survival to levels comparable to the M1 subsets. Instead, VitD3-polarized cells very efficiently controlled Mtb infection over time. Despite a higher uptake of Mtb bacilli in VitD3-polarized compared to M1 cells (33% versus 18%), Mtb growth of both H37Rv and an MDR-TB isolate was maintained at stable, low levels at 24 and 72 h post-Mtb infection. In addition, Mtb growth was around 40% lower in VitD3-polarized macrophages compared with the unstimulated M0 control. A previous study reported that macrophages treated with a low-dose of VitD3 during differentiation, were less infected with dengue virus when compared to untreated macrophages (Arboleda Alzate, Rodenhuis-Zybert et al. 2017). Our experiments
infection more efficiently compared to M1- or M2-polarized cells. Thus, sufficient VitD3
status may be crucial to support macrophage polarization in vivo.
Normally, M1 macrophages have been found to be better equipped with microbial killing capacities (Flannagan, Cosio et al. 2009). It was previously demonstrated that M1 macrophages are infected with BCG at a lower proportion as compared to M2 macrophages, and similar levels of BCG infection were maintained even at 6 days post-infection. (Verreck, de Boer et al. 2004). Our results of reduced infectivity of M1 compared to M2 macrophages at 4 h concur with their findings, but at 3 days post-infection we observed progressive multiplication of bacteria, which could be a trait of intracellular survival or persistence in virulent mycobacteria. Similarly, it has been shown in a chlamydia infection model that despite a lower infectivity of M1 macrophages, bacteria persist in the M1 subset while permissive growth is evident in M2 macrophages (Gracey, Lin et al. 2013). These findings support the notion that several pathogens have developed strategies to persist for prolonged periods of time in the host and may allow low, but persistent growth in activated macrophages despite induction of antimicrobial responses.
3.4.4 Mtb-infected VitD3-polarized macrophages are more pro-inflammatory and express higher mRNA levels of LL-37 compared with M1 and M2 subsets
To obtain a better view on the function of the different macrophage subsets, we quantified host cell mRNA expression of pro-inflammatory, antimicrobial as well as inhibitory effector molecules before and after Mtb infection. VitD3-polarized macrophages specifically up-regulated TNFα, IL-1β, CCL2, and LL-37, whereas immunosuppressive molecules such as IL-1RA, IDO and Arg-1 were expressed at relatively lower levels after Mtb infection. VitD3-polarized cells also expressed relatively high levels of IL-10 and LC3b compared with the M1 and M2 subsets, but this expression was not changed with Mtb infection. Impaired TNFα and IL-1β production, may worsen disease progression in TB patients (Waitt, Banda et al. 2015), while CCL2 and TNFα enhance recruitment of immune cells to the site of Mtb infection (Hasan, Cliff et al. 2009). As we observed in Study I, Mtb infection resulted in a potent down-regulation of LL-37, however, LL-37 expression was still substantially higher in Mtb-infected VitD3-polarized cells compared to M1 and M2 subsets. As mentioned above, TB patients expressed low levels of LL-37 in TB granulomas (Rahman, Rehn et al. 2015), while VitD3 supplementation enhanced the production of LL-37 in blood cells (Mily, Rekha et al. 2015). Contrary to VitD3-polarized
levels of LL-37. M2-polarized macrophages also expressed high IL-10 levels, but this was not changed with Mtb infection. As expected, the ratio between iNOS/Arg-1 expression was higher in M1- compared with M2-polarized macrophages (Flynn, Chan et al. 2011).
The immunosuppressive enzyme IDO, effectively diminish activation of Th1 cells by degradation of the essential amino acid tryptophan that may result in growth arrest of T cells as well as decreased activity of antigen presenting cells (APCs) (Harden and Egilmez 2012).
As such, enhanced IDO activity in PBMCs from individuals vaccinated with a novel TB vaccine candidate was inversely correlated to CD4+IFNγ+ T cell responses (Tanner, Kakalacheva et al. 2014). Increased IDO expression has been detected in the macrophage rich areas of TB granulomas (Mehra, Alvarez et al. 2013) and high levels of IDO has also been found in patients with TB pleuritis (Suzuki, Miwa et al. 2013). Interestingly, inhibition of IDO in Mtb-infected non-human primates enhanced T cell proliferation and recruitment of effector T cells to the granulomas, which resulted in enhanced bacterial control (Gautam, Foreman et al. 2018). It has been reported that VitD3 possess anti-inflammatory properties, which may protect the host from extensive tissue damage and inflammation (Harishankar, Anbalagan et al. 2016). Thus, VitD3 may enhance innate antimicrobial responses but simultaneously contribute to responses that dampens inflammation but prevents immunosuppression. Further in vitro and in vivo studies will enhance our understanding of the role of VitD3 on macrophage polarization and protective TB immunity.